update llm infer docs #9314
Merged
update llm infer docs #9314
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #9314 +/- ##
===========================================
- Coverage 53.44% 52.41% -1.04%
===========================================
Files 664 661 -3
Lines 109935 108376 -1559
===========================================
- Hits 58757 56801 -1956
- Misses 51178 51575 +397 ☔ View full report in Codecov by Sentry. |
ZHUI
reviewed
Oct 24, 2024
llm/docs/predict/inference.md
Outdated
| @@ -94,6 +95,8 @@ PaddleNLP 提供了多种参数，用于配置推理模型和优化推理性能 | |||
|
|
|||
| - `block_attn`: 是否使用 Block Attention 推理， 默认值为False。Block Attention 是基于 PageAttention 的思想设计并实现的，在保持高性能推理和动态插入的基础上可以动态地为 cachekv 分配存储空间，极大地节省显存并提升推理的吞吐。 | |||
|
|
|||
| - `append_attn`: Append Attention 在 Block Attention 实现的基础上，进一步借鉴 FlashInfer 的实现对 Attention 模块进行了优化，并增加了C4的高性能支持，极大地提升了推理性能。 | |||
There was a problem hiding this comment.
最好说明一下两者的关系，是二选一，还是叠加？
建议可以说一下 append_attn 的优势是什么？哪些场景下更合适，这里只说借鉴 FlashInfer，但是用户可能不知道FlashInfer是什么。
There was a problem hiding this comment.
二选一，全场景下应该都合适，属于是block_attn的升级版
| @@ -43,27 +43,27 @@ BF16推理 | |||
|
|
|||
| ```shell | |||
| # 动态图推理 | |||
| python ./predict/predictor.py --model_name_or_path meta-llama/Meta-Llama-3-8B-Instruct --dtype bfloat16 --mode dynamic --inference_model 1 --block_attn 1 | |||
| python ./predict/predictor.py --model_name_or_path meta-llama/Meta-Llama-3-8B-Instruct --dtype bfloat16 --mode dynamic --inference_model 1 --append_attn 1 | |||
There was a problem hiding this comment.
文档是直接对外，append_attn 相关功能的CI也辛苦加一下，避免用户跑不通的情况。
There was a problem hiding this comment.
CI有些问题，已经在调试了，在下个PR中会提交
qingqing01
approved these changes
Oct 25, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
Others
PR changes
Docs
Description
update llm infer docs